The ability to flexibly adapt behavior based on current goals and context is called cognitive control, which is essential for responding appropriately to the diverse situations we face in life. This becomes strikingly clear when cognitive control is impaired, as in schizophrenia, obsessive-compulsive disorder and attention-deficit hyperactivity disorder. An important implementation of cognitive control is the application of rules, which map cues to actions according to context. When we carry out actions, we often start by applying abstract rules (e.g. morning means coffee), which aid in the selection of more concrete rules (e.g. coffee means grind beans) closer to action specification. Prefrontal cortex (PFC), especially areas 46 and 9/46, is vital for processing rules, and PFC neurons have been shown to represent concrete and abstract rules. This raises two fundamental questions. First, how are neurons that represent abstract and concrete rules organized in PFC? Although there are two prominent proposals for the functional organization of PFC, one suggesting an anterior- posterior gradient based on rule abstraction and the other suggesting individual neurons contribute to the processing of multiple rules (instead of being topographically organized), there is a lack of electrophysiology studies testing these proposals. The second question is how are distinct ensembles of PFC neurons flexibly and selectively activated based on behaviorally relevant rules? Neural synchrony may be a suitable selection mechanism, dynamically routing information between synchronized cells. Evidence suggests that the higher- order thalamus, which forms indirect pathways between cortical areas, can regulate cortical oscillations and synchrony. The mediodorsal thalamic nucleus (MD) is extensively connected with PFC and is thus well positioned to influence PFC activity. Functional MRI and lesion studies suggest that MD plays an important role in rule processing and cognitive control in general. However, there have been very few electrophysiology studies probing the role of MD in cognitive control, and none have probed how MD and PFC interact in primates. The goal of the proposed research is to characterize how MD contributes to rule processing (SA#1), how PFC is functionally organized (SA#2), and how MD and PFC interact during rule-guided behavior (SA#3). The central hypothesis is that MD regulates information transmission between PFC neurons. A key mechanism may involve MD synchronizing PFC neurons that represent task-relevant rules. To test this hypothesis, we simultaneously record neural activity in MD and areas 46 and 9/46 of monkeys performing a rule-based task. We also stimulate MD to test whether MD has a causal influence on oscillatory activity, neural synchrony and information transmission across PFC. To translate this work to humans, we acquire intracranial recordings in epilepsy patients performing the same rule-based task. The proposed research will advance our understanding of the large-scale network dynamics that mediate cognitive control. Defining the basic mechanisms of cognitive control is a first necessary step in developing effective treatment strategies for cognitive control deficits.